Not Just Software vs. Hardware
We recently took a look at Lattice’s approach to sensor hubs. We’ve seen many other ways of implementing sensor hubs in the past, but all of those were software-based; it was just a question of where the software executes. Lattice’s approach is hardware, and that raises all kinds of new questions.
The biggest red flag that it raises for me is that moving a task from software to hardware in the design phase is not trivial. (Trying to keep it with the software guys, using tools that automatically generate hardware is, for the most part, a quixotic goal that seems largely to have been lovingly placed back on the shelf.) In my quest to figure this part out, I found that there’s more to the sensor hub world than all-software and all-hardware. And that makes the design question even more complex.
Indoor Location System Knows Where Your Treasure is Buried
As first-world problems go, losing your car keys is a bad one. Losing a whole warehouse full of shippable merchandise is probably worse, but warehouses typically have lots of people standing around watching over the goods. But what do you do when you’ve lost your car keys somewhere inside the warehouse? Needle, meet haystack. Not so easy now, huh? What are you going to do?
If you’ve planned ahead, you’d have little badges on your keys – badges stuffed with technology designed by DecaWave. The Dublin, Ireland startup has created an itty-bitty little chip it calls the ScenSor that’s designed to solve this, and much larger, problems.
Engaging More Senses
I did a lot of typing in college. Even as an engineering student, I still had to take humanities courses, which meant writing papers. And, as cash was in short supply, I found that I could leverage this skill by typing papers for others.
I had an electric typewriter (no, I’m not that old) and the special keys for Spanish, French, and German, making me versatile. And so I got pretty good at blind touch-typing. In other words, I would simply look at the paper to be typed and proceed without glancing back at the results. In fact, I found that I could get into a zone where my fingers would be blazing away with no mistakes – that is, until I realized that and started thinking, “Wow, this is going really well,” at which point it started falling apart as my brain started to try to micromanage my fingers (like the centipede trying to think through how it can walk).
Embedded Software Store, DigitalOptics, and Intelligent Robots
In honor of Thanksgiving next week, we're serving up a bountiful virtual turducken* of EE goodness. For the first course, we tackle that age old question: How do you train a fleet of robots to obey your every command? (The key to this quandary may be as close as the eyes on our faces.) In our next course, we serve up a feast of embedded software and embedded IP with Sheneka Coleman from Avnet. Lastly, we finsh up our fine Friday Fun meal with a tasty morsel from Eric Sigler of DigitalOptics. (DigitaOptics's new mems|cam may change how we interact with our smartphones forever.) This Thankgiving fryer won't be hot for too much longer, so dig in while the gettin's good.
MathWorks Simplifies the Standard
The mobile world is undergoing a wonderful standards convergence. Instead of an alphabet soup of competing standards - GSM, CDMA, HSPA, HSDPA, OFDM, WiMax, EDGE, UMTS, and a host of alternatives and variants spreading their spectrum around the world and subdividing our engineering efforts into an unmanageably complicated array of special cases - we now have LTE (for “Long Term Evolution”) which has won favor both across a wide swath of the industry geographically and as a standard with some legs that may keep it around longer than two generations of smartphones.
Before there was LTE, the market was badly fragmented. Just about every geographical region had its own standard or version of a standard. Interoperability was weak to nonexistent, and getting economy of scale on engineering efforts and investment was near impossible. In that world, each project became a one-off, with few tools and little IP that were much use in the specialized task of designing and verifying a compliant device or system.
2013 MEMS Executive Congress
MEMS. The final frontier. What better place to check out some of the coolest innovation in our electronic engineering community than the 2013 MEMS Executive Congress. QuickLogic has jumped into the MEMS space in big way, and this week Brian Faith (VP - QuickLogic) breaks it all down for us. We're also waxing MEMS poetic with Alissa Fitzgerald (Founder - AMFitzgerald) and Peter Himes (VP - Silex) about the future of innovation in the MEMS sphere. Join us as we get our MEMS on!
QuickLogic Announces Ultra-Low-Power Sensor Hub
Sensors are literally taking over the world. Projections vary as to actual numbers - some say we will reach a trillion - but it is safe to say that we are in the middle of an exponential explosion in the number of sensors deployed in the world. Beyond the obvious gajillions in smartphones, sensors are being designed into just about every kind of embedded system you can imagine. All those sensors promise a revolution in the real-world intelligence of the systems we all design.
One of the biggest problems with putting sensors into our systems is - they’re kinda high-maintenance from a processing point of view. All day long, like cranky little toddlers, they keep dribbling out data and wanting us to deal with it. If you get very many sensors in your system, you really need a nanny to take care of them all, so you can get back to the important business of letting your embedded computer do some embedded computing.
New Low-end x86 MCU Line Emerging from the Lab
Intel is going subatomic in its charge to empower embedded systems.
Last week the company teased out a few hints about a new low-end product line called Quark. Quark is like Atom, but smaller. (Get it? Quark? Atom? It’s nerdy.) Like Atom, Quark chips will be x86-based embedded processors, but, unlike Atom, they’ll be more like an MCU or SoC. Indeed, the first (and so far, only) Quark chip revealed has “SoC” right in its name.
Like real quarks, the Intel version is hard to capture and pin down. The company has revealed almost nothing about Quark’s features or technical merits, preferring instead to just telegraph its intention to enter the market for “intelligent systems” that need x86 compatibility.
Nirvanix Failure Leaves Customers in the Lurch
Let me just start by saying, “I Told You So.”
Cloud-storage company Nirvanix just filed for Chapter 11 bankruptcy protection, abruptly shutting its doors and informing its customers that they have exactly two weeks to come get their data.
Two whole weeks. For every customer to retrieve all the data that was presumably (a) massive enough to require offsite cloud storage, and (b) important enough to require offsite cloud storage.
What are you going to do, start a marathon download from Nirvanix’s servers to yours? And then upload it all again to a different service? Hope you've got a really fast connection at your office. Alternatively, you’ve got two weeks to hustle your butt down to Nirvanix’s San Diego headquarters with a bunch of DVDs, tapes, or USB drives under your arm.
Movea and Hillcrest Labs in Agreement?
This was going to be a story exposing a simmering feud, an ongoing debate, a religious war. As it turns out, it feels more like violent agreement than anything else.
I’m not here to point fingers at anyone. I am here, however, to discuss pointing. I’ll bet you didn’t know there was a technical debate about that, now, did you? Let’s talk first about what we’re talking about, and then I’ll return to the controversial (?) bits.
Pointing is, of course, natural. Even dogs can figure out when we’re pointing at something. (Some dogs, anyway… others prefer to keep looking at your finger…) And so we’ve worked it into our technology, ever since Apple Xerox came up with the concept of the mouse.